insitro, a leading machine learning–driven drug discovery and development company, has announced the next phase of its ongoing collaboration with Bristol Myers Squibb (NYSE: BMY) to develop novel therapeutics for amyotrophic lateral sclerosis (ALS). The collaboration extension deepens the partnership’s focus on leveraging insitro’s ChemML™ platform, an AI-enabled end-to-end drug design system, to translate promising biological discoveries into viable small-molecule medicines.
This one-year extension may provide up to $20 million in new funding, building upon a successful first phase that identified multiple novel ALS targets through insitro’s integrated biology and machine learning approaches. If the partnership yields a successful therapeutic, the total potential value to insitro could exceed $2 billion, including discovery, development, regulatory, and commercial milestone payments, plus royalties on future sales.
Turning Biological Discoveries into Therapeutic Breakthroughs
“Our collaboration with Bristol Myers Squibb has uncovered novel targets with the potential to address the underlying biology of ALS,” said Daphne Koller, Ph.D., founder and CEO of insitro. “We are now entering the next phase—turning these discoveries into medicines. With ChemML, our end-to-end drug design platform, we can translate novel targets into advanced small-molecule leads rapidly, leveraging a differentiated set of capabilities that span AI-driven modeling, medicinal chemistry, and structural biology.”
Koller emphasized that insitro’s goal remains focused on developing transformative, disease-modifying therapies for ALS. “While advancing these initial drug candidates, we will continue our efforts to identify additional new targets. Our aim remains unwavering: to deliver truly transformative treatments that enable people with ALS to live longer,” she said.
Harnessing ChemML™ for End-to-End AI-Powered Drug Discovery
insitro’s ChemML™ platform represents a breakthrough in small-molecule discovery, uniting machine learning, laboratory automation, and advanced simulation to accelerate every step of the drug design process. Built through internal development and enhanced by the acquisition of Haystack Sciences, the ChemML™ platform is designed to streamline drug creation for complex and difficult-to-treat diseases.
While the pharmaceutical industry has made strides in applying AI and ML to drug discovery, progress has often been incremental and costly. ChemML™ addresses these challenges by tightly integrating in silico modeling with real-world experimentation, enabling a continuous feedback loop that enhances accuracy and efficiency over time.
Key features of ChemML™ include:
- Data generation at scale:
insitro’s proprietary Quantitative Adaptive Libraries (QALs) can produce hundreds of millions of data points related to drug–target binding and selectivity. This enables the creation of highly predictive ML models and supports rapid, data-driven discovery cycles. - Predictive pharmacological property modeling:
Advanced ML models trained on high-quality datasets—developed in part through insitro’s collaboration with Eli Lilly and Company—accurately predict ADMET (absorption, distribution, metabolism, excretion, and toxicity) properties, including in vivo pharmacokinetics. - AI-driven iterative design loop:
A proprietary design-make-test engine continuously improves with each iteration, combining computational predictions and experimental results to guide compound synthesis more intelligently and efficiently. - Powerful compute infrastructure:
insitro operates a high-performance compute cluster featuring 192 H100 GPUs, enabling cutting-edge modeling, structural prediction, and physics-based simulation at scale.
Accelerating ALS Drug Discovery Through AI
“insitro’s machine learning–enabled discovery engine has revealed new biology that allowed us to identify multiple differentiated, high-confidence ALS drug targets at record speed,” said Philip Tagari, Chief Scientific Officer of insitro.
“These targets are supported by robust evidence, including functional data showing improvements in motor neuron survival and reversal of multiple downstream markers of ALS pathology,” Tagari added. “Our ChemML™ platform has already generated novel chemical compounds that can now be optimized through our ML-powered medicinal chemistry to address technically challenging targets.”
The collaboration between insitro and Bristol Myers Squibb aims to accelerate ALS drug discovery by combining insitro’s AI-driven precision with Bristol Myers Squibb’s deep expertise in clinical development and neuroscience. The partners share a commitment to bringing meaningful new options to ALS patients, a group that has long faced limited therapeutic choices.
Targeting ALS: A Critical Unmet Need
ALS is a progressive neurodegenerative disorder that leads to the degeneration of upper and lower motor neurons, causing muscle weakness, respiratory failure, and eventually death. The median survival after diagnosis is typically three to five years. Despite decades of research, effective therapies that significantly slow or halt disease progression remain elusive.
Nearly 90% of ALS cases are sporadic, meaning they occur without a clear familial or genetic cause. For this reason, the first phase of the insitro–Bristol Myers Squibb collaboration focused on identifying cross-cutting disease biology—the shared molecular mechanisms underlying both familial and sporadic ALS. This approach has the potential to yield therapies that can benefit a broader population of patients.
Both companies are dedicated to advancing their joint research at pace, recognizing the urgency of the need. “We are working with urgency and precision,” said Koller. “Each insight generated through ChemML™ brings us closer to delivering therapies that could make a real difference for ALS patients and their families.”
Building the Future of AI-Enabled Drug Discovery
The extension of the collaboration marks an important milestone for both insitro and Bristol Myers Squibb. It demonstrates how AI and machine learning are reshaping the landscape of pharmaceutical innovation, enabling faster, more data-informed decision-making and a more efficient path from discovery to clinic.
By harnessing computational power and biological insight, insitro aims to bridge the gap between genetic discoveries and therapeutic solutions, reducing the time and cost required to move from lab to patient. The company’s growing portfolio of partnerships reflects increasing recognition across the biopharma industry of the value of AI-native discovery platforms.
“Our partnership with Bristol Myers Squibb exemplifies what’s possible when cutting-edge data science meets world-class drug development,” said Koller. “We look forward to continuing our shared journey to transform how we understand—and ultimately treat—complex diseases like ALS.”
